platform/upstream/efl.git
9 years agoEvas GL: Fix list of GLES 1.1 extensions
Jean-Philippe Andre [Fri, 19 Dec 2014 09:06:23 +0000 (18:06 +0900)]
Evas GL: Fix list of GLES 1.1 extensions

We should use GLESv1 functions in a GLESv1 context to scan for
GLESv1 extensions. Makes sense yeah?

This should expose the proper list... especially enabling FBO
extension when it's supported by the driver.

9 years agoEvas GL: Add a warning in gl_x11 about MSAA & RGBA for GLES1
Jean-Philippe Andre [Fri, 19 Dec 2014 08:36:14 +0000 (17:36 +0900)]
Evas GL: Add a warning in gl_x11 about MSAA & RGBA for GLES1

We've discovered some bugs on some drivers related to
MSAA and RGBA blending. Dunno yet if the same problem can affect
GLES2 as well.

9 years agoelua: getopt updates
Daniel Kolesa [Thu, 18 Dec 2014 16:09:58 +0000 (16:09 +0000)]
elua: getopt updates

9 years agoedje: fix memory leak in epp
Thiep Ha [Thu, 18 Dec 2014 00:17:03 +0000 (09:17 +0900)]
edje: fix memory leak in epp

Summary:
Variable include_defaults can be allocated and
go out of scope without freeing it.

Reviewers: cedric, seoz, woohyun, Hermet

Reviewed By: Hermet

Differential Revision: https://phab.enlightenment.org/D1784

9 years agoedje/cpplib: fix formatting
ChunEon Park [Thu, 18 Dec 2014 00:09:33 +0000 (09:09 +0900)]
edje/cpplib: fix formatting

9 years agoedje: free allocated memory in epp
Thiep Ha [Thu, 18 Dec 2014 00:05:50 +0000 (09:05 +0900)]
edje: free allocated memory in epp

Summary: fname variable's memory is allocated but not freed.

Reviewers: cedric, seoz, woohyun, Hermet

Reviewed By: Hermet

Differential Revision: https://phab.enlightenment.org/D1785

9 years agoeldbus: Fix crash when removing the last reference of the message container inside...
José Roberto de Souza [Wed, 17 Dec 2014 18:57:03 +0000 (16:57 -0200)]
eldbus: Fix crash when removing the last reference of the message container inside of the message callback

If user try to remove the last reference of proxy, object, connection
or eldbus(lib) inside of message callback it was causing the
eldbus_pending_dispatch() being called 2 times, one because of the
eldbus_cancel() that is triggered when the last reference of the
message parent is removed and another after the return of the user
callback.

==6545== Invalid read of size 8
==6545==    at 0x52F784E: eldbus_cbs_free_dispatch (eldbus_core.c:266)
==6545==    by 0x53064AA: eldbus_pending_dispatch (eldbus_pending.c:227)
==6545==    by 0x5305961: cb_pending (eldbus_pending.c:74)
==6545==    by 0x6B29DB1: ??? (in /usr/lib/libdbus-1.so.3.8.9)
==6545==    by 0x6B2D280: dbus_connection_dispatch (in /usr/lib/libdbus-1.so.3.8.9)
==6545==    by 0x52F93B4: eldbus_idler (eldbus_core.c:773)
==6545==    by 0x4E4B300: _ecore_call_task_cb (ecore_private.h:305)
==6545==    by 0x4E4B78F: _ecore_idler_all_call (ecore_idler.c:143)
==6545==    by 0x4E4EA73: _ecore_main_loop_spin_core (ecore_main.c:1768)
==6545==    by 0x4E4EAF1: _ecore_main_loop_spin_timers (ecore_main.c:1802)
==6545==    by 0x4E4ED01: _ecore_main_loop_iterate_internal (ecore_main.c:1925)
==6545==    by 0x4E4D03B: ecore_main_loop_begin (ecore_main.c:983)
==6545==  Address 0x701aa78 is 104 bytes inside a block of size 128 free'd
==6545==    at 0x4C2B200: free (in /usr/lib/valgrind/vgpreload_memcheck-amd64-linux.so)
==6545==    by 0x530655B: eldbus_pending_dispatch (eldbus_pending.c:241)
==6545==    by 0x5306763: eldbus_pending_cancel (eldbus_pending.c:259)
==6545==    by 0x52F29DB: _eldbus_proxy_clear (eldbus_proxy.c:146)
==6545==    by 0x52F3057: _eldbus_proxy_unref (eldbus_proxy.c:244)
==6545==    by 0x52F3393: eldbus_proxy_unref (eldbus_proxy.c:264)
==6545==    by 0x401039: on_get_playlists (banshee.c:53)
==6545==    by 0x5306493: eldbus_pending_dispatch (eldbus_pending.c:225)
==6545==    by 0x5305961: cb_pending (eldbus_pending.c:74)
==6545==    by 0x6B29DB1: ??? (in /usr/lib/libdbus-1.so.3.8.9)
==6545==    by 0x6B2D280: dbus_connection_dispatch (in /usr/lib/libdbus-1.so.3.8.9)
==6545==    by 0x52F93B4: eldbus_idler (eldbus_core.c:773)

Now we will remove the pending from parent pending list before
call the user callback, this way only the pending messages will
be canceled.

Also we need increase the eldbus reference before call
dbus_connection_dispatch() or user could remove the last reference of
eldbus inside of a message callback when we still are
holding one reference of the connection.

@fix
ref T1908

9 years agoRevert "eldbus hackaround for proxy deletion during pending callback"
José Roberto de Souza [Wed, 17 Dec 2014 18:44:29 +0000 (16:44 -0200)]
Revert "eldbus hackaround for proxy deletion during pending callback"

This reverts commit 6a08788d59c020aea8e0f24dc072630d5ae61c18.

This will be properly fixed on the next patch.

9 years agoelua: document getopt.lua
Daniel Kolesa [Wed, 17 Dec 2014 17:09:47 +0000 (17:09 +0000)]
elua: document getopt.lua

9 years agoelua: error reporting in the library
Daniel Kolesa [Wed, 17 Dec 2014 14:43:50 +0000 (14:43 +0000)]
elua: error reporting in the library

9 years agoevas - generic loaders - fix possible string overflow on decoders
Jaehyun Cho [Wed, 17 Dec 2014 08:48:51 +0000 (17:48 +0900)]
evas - generic loaders - fix possible string overflow on decoders

Summary:
fix possible string overflow on decoders.
         this commit is based on commit 3c5412f72879bbce72bfbc7190acd9e5d16d3b46
         that fixes CID 1039580.

Reviewers: raster

Reviewed By: raster

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1783

9 years agoeo: Fix example build problem with DSO.
Daniel Juyung Seo [Wed, 17 Dec 2014 07:58:23 +0000 (16:58 +0900)]
eo: Fix example build problem with DSO.

/usr/bin/ld: isa/eo_isa_interface.o: undefined reference to symbol 'eina_main_loop_is'
/usr/bin/ld: note: 'eina_main_loop_is' is defined in DSO /home//seoz/efl/git/usrcore//binefl//ldsrc:/ libsimple//einasimple_interface.o/:. libsundefined/ libeina.so.1reference  soto  trysymbol  adding' eina_main_loop_isit'
to/ usrthe/ binlinker/ ldcommand:  linenote

9 years agoecore animator pos map - move from v[] array to *v pointer.
Carsten Haitzler (Rasterman) [Wed, 17 Dec 2014 08:00:11 +0000 (17:00 +0900)]
ecore animator pos map - move from v[] array to *v pointer.

this moves from double v[][ as a param to double *v - which is more
consistent with the rest of efl. this shouldnt be an abi break.

9 years agoevas/tgv: Added DSSIM metric in evas_image_save_tgv
kabeer khan [Wed, 17 Dec 2014 07:40:59 +0000 (16:40 +0900)]
evas/tgv: Added DSSIM metric in evas_image_save_tgv

Summary:
Resolved TODO to add DSSIM metric in DEBUG_STATS

Signed-off-by: kabeer khan <kabeer.khan@samsung.com>
Reviewers: cedric

Subscribers: devilhorns, cedric

Differential Revision: https://phab.enlightenment.org/D1773

9 years agoUse NEON intrinsics for mapping instead of inline asm
Carsten Haitzler (Rasterman) [Wed, 17 Dec 2014 06:28:50 +0000 (15:28 +0900)]
Use NEON intrinsics for mapping instead of inline asm

Summary: Rewrite linline assembly in mapping func using NEON intrinsics.

Reviewers: raster

Differential Revision: https://phab.enlightenment.org/D1740

9 years agoecore-drm: add error handling, when fail to setup tty.
Seunghun Lee [Tue, 16 Dec 2014 14:20:25 +0000 (09:20 -0500)]
ecore-drm: add error handling, when fail to setup tty.

Summary: need to restore vt to text mode, when failed to setup.

@fix

Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1777

9 years agoLogically dead code has been removed.Because width will always be less than HOST_BITS...
Nibha Sharma [Tue, 16 Dec 2014 11:08:31 +0000 (20:08 +0900)]
Logically dead code has been removed.Because width will always be less than HOST_BITS_PER_INT so else branch will never execute.

Reviewers: singh.amitesh, seoz, cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1506

9 years agoecore: removed switch case 0 condition
Nibha Sharma [Tue, 16 Dec 2014 10:59:39 +0000 (19:59 +0900)]
ecore: removed switch case 0 condition

Summary: It is removed because it was a deadcode case zero is never reached and executed.

Reviewers: singh.amitesh, seoz, raster

Reviewed By: raster

Subscribers: raster, cedric

Differential Revision: https://phab.enlightenment.org/D1432

9 years agoecore_con: Added NULL check in dns_so_reset() for dns_socket object.
Amitesh Singh [Tue, 16 Dec 2014 10:55:12 +0000 (19:55 +0900)]
ecore_con: Added NULL check in dns_so_reset() for dns_socket object.

Summary: @fix

Reviewers: zmike, tasn, seoz, raster

Subscribers: cedric, seoz

Differential Revision: https://phab.enlightenment.org/D1411

9 years agoEina streq: remove redundant code in eina_streq.
Tom Hacohen [Tue, 16 Dec 2014 09:23:43 +0000 (09:23 +0000)]
Eina streq: remove redundant code in eina_streq.

9 years ago+eina_memdup
Mike Blumenkrantz [Mon, 15 Dec 2014 19:30:43 +0000 (14:30 -0500)]
+eina_memdup

for those times when three lines of code should be one

@feature

9 years agoeina_streq should immediately return true if both pointers are the same
Mike Blumenkrantz [Mon, 15 Dec 2014 19:21:44 +0000 (14:21 -0500)]
eina_streq should immediately return true if both pointers are the same

9 years agoEvas textblock: Remove redundant magic member.
Tom Hacohen [Mon, 15 Dec 2014 14:58:32 +0000 (14:58 +0000)]
Evas textblock: Remove redundant magic member.

We don't use magic since moving to Eo.

9 years agoecore: Added getting "WarningLevel" on upower module.
MinJeong Kim [Fri, 12 Dec 2014 12:09:09 +0000 (13:09 +0100)]
ecore: Added getting "WarningLevel" on upower module.

Summary:
From UPower 0.99.0, a property "OnLowBattery" was removed and
upower have recommended to use "WarningLevel" instead.
This revision provides "WarningLevel" using the property "DaemonVersion"
and it keeps dealing with "OnLowBattery" as well for old system.

Fixes T1909

Reviewers: gwanglim, barbieri, zmike, stefan_schmidt

Reviewed By: barbieri

Subscribers: zmike, stefan_schmidt, cedric

Differential Revision: https://phab.enlightenment.org/D1717

9 years agoedje: Removed redundant assignement of rest_args.
Srivardhan Hebbar [Fri, 12 Dec 2014 14:14:00 +0000 (09:14 -0500)]
edje: Removed redundant assignement of rest_args.

Summary:
While going through the code, I found this. Thought this was just a redundant and fixed it.

@fix

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1774

9 years agoelua: remove main.h (launcher is only 1 C file now)
Daniel Kolesa [Fri, 12 Dec 2014 13:43:39 +0000 (13:43 +0000)]
elua: remove main.h (launcher is only 1 C file now)

9 years agoelua: move caching into the library
Daniel Kolesa [Fri, 12 Dec 2014 13:39:57 +0000 (13:39 +0000)]
elua: move caching into the library

9 years agoelua: move io extensions to library
Daniel Kolesa [Fri, 12 Dec 2014 12:04:10 +0000 (12:04 +0000)]
elua: move io extensions to library

9 years agoelua: move i18n setup to the library
Daniel Kolesa [Fri, 12 Dec 2014 10:44:58 +0000 (10:44 +0000)]
elua: move i18n setup to the library

9 years agoedje: Edje_Edit - add generation of 'filter' parameter of program
Kateryna Fesyna [Fri, 12 Dec 2014 03:37:41 +0000 (04:37 +0100)]
edje: Edje_Edit - add generation of 'filter' parameter of program

Reviewers: raster, Hermet, cedric

@feature

Reviewed By: cedric

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D1749

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoedje: Edje_Edit - add API for program actions PLAY_SAMPLE and PLAY_TONE
Andrii Kroitor [Fri, 12 Dec 2014 03:33:26 +0000 (04:33 +0100)]
edje: Edje_Edit - add API for program actions PLAY_SAMPLE and PLAY_TONE

Summary:
Added initialisation, deletion, getters and setters for following
programs fields: sample_name, speed, channel, tone_name, duration.

@feature

Reviewers: cedric, seoz, Hermet

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D1753

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoedje: use realloc instead of malloc and memcpy.
Srivardhan Hebbar [Fri, 12 Dec 2014 03:22:49 +0000 (04:22 +0100)]
edje: use realloc instead of malloc and memcpy.

Summary:
Replaced malloc with realloc. Removed free. Added a Error message.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: cedric

Reviewed By: cedric

Subscribers: devilhorns, cedric

Differential Revision: https://phab.enlightenment.org/D1766

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoedje: Edje_Edit - generate alias parameter for 'parts' block.
Maksym Volodin [Fri, 12 Dec 2014 03:20:16 +0000 (04:20 +0100)]
edje: Edje_Edit - generate alias parameter for 'parts' block.

Summary: add alias parameter to code generation.

@feature

Reviewers: raster, Hermet, cedric

Reviewed By: cedric

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D1768

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore_ipc: coverity issue - you cannot return without free
Shinwoo Kim [Fri, 12 Dec 2014 03:05:36 +0000 (04:05 +0100)]
ecore_ipc: coverity issue - you cannot return without free

Summary: The allocated memory is not released before return.

Lost track of the CID.

Test Plan: Run static analysis tool such as prevent

Reviewers: raster, cedric

Reviewed By: cedric

Subscribers: cedric, seoz

Differential Revision: https://phab.enlightenment.org/D1746

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years ago.gitignore: ignore generated files.
Cedric BAIL [Fri, 12 Dec 2014 02:46:36 +0000 (03:46 +0100)]
.gitignore: ignore generated files.

9 years agoelua: fix make helper
Daniel Kolesa [Thu, 11 Dec 2014 17:11:30 +0000 (17:11 +0000)]
elua: fix make helper

9 years agoelua: move core scripts to src/scripts
Daniel Kolesa [Thu, 11 Dec 2014 16:59:27 +0000 (16:59 +0000)]
elua: move core scripts to src/scripts

9 years agoelua: add checkme
Daniel Kolesa [Thu, 11 Dec 2014 16:41:24 +0000 (16:41 +0000)]
elua: add checkme

9 years agoRevert "Evas: Stop installing Evas_Engine_Stuff.h"
Chris Michael [Thu, 11 Dec 2014 16:10:56 +0000 (11:10 -0500)]
Revert "Evas: Stop installing Evas_Engine_Stuff.h"

This reverts commit df3958bb892b313ab91b9f0bd5a84452bb62388c.

This is getting reverted because it broke building of Enlightenment. E
requires the Evas_Engine_Buffer header file for drawing mouse cursor
using the buffer engine.

9 years agoelua: add initial pieces of the library
Daniel Kolesa [Thu, 11 Dec 2014 15:48:18 +0000 (15:48 +0000)]
elua: add initial pieces of the library

9 years agoREADME: Remove outdated version number
Stefan Schmidt [Thu, 11 Dec 2014 13:07:15 +0000 (14:07 +0100)]
README: Remove outdated version number

We have the version in configure.ac and NEWS where it gets updated for
releases. This occurrence gets forgotten to often on the other hand. Its
not needed here so we remove it.

9 years agoEvas: Stop installing Evas_Engine_Stuff.h
Jean-Philippe Andre [Wed, 10 Dec 2014 11:26:55 +0000 (20:26 +0900)]
Evas: Stop installing Evas_Engine_Stuff.h

Those really are internals shared between ecore and evas.
Considering ecore & evas are just part of EFL, and expedite
now doesn't even rely on these headers anymore, we can safely
remove them from the dist packages.

This will allow us to break this seemingly internal API/ABI.

9 years agoecore-drm: Free devices before we release the libinput reference
Chris Michael [Wed, 10 Dec 2014 19:26:52 +0000 (14:26 -0500)]
ecore-drm: Free devices before we release the libinput reference

Summary: When destroying evdev devices, we make calls to
libinput_device_unref ... but we cannot Make that call if the libinput
context has been destroyed already (as happens when we free input). So
to fix, don't release the libinput context until all devices have been
released.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoevas-gl-common: Fix issue with Coverity reporting null dereferences
Chris Michael [Wed, 10 Dec 2014 18:30:51 +0000 (13:30 -0500)]
evas-gl-common: Fix issue with Coverity reporting null dereferences

Summary: This fixes Coverity CID1257606 and CID1257607: Dereferencing
null return value. _evgl_tls_resource_get Can return NULL so we should
be checking that returned value before trying to use it

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeldbus: Fix possible null dereference reported by Coverity
Chris Michael [Wed, 10 Dec 2014 18:24:54 +0000 (13:24 -0500)]
eldbus: Fix possible null dereference reported by Coverity

Summary: Fix Coverity CID1256952: reports a null derefence here due to
eldbus_message_new returning NULL, thus causing a null dereference
when trying to set reply->dbus_msg

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Don't declare sprite variable unless we need it, also fix
Chris Michael [Wed, 10 Dec 2014 17:17:37 +0000 (12:17 -0500)]
ecore-drm: Don't declare sprite variable unless we need it, also fix
some formatting.

Summary: This commit saves from declaring variable for
Ecore_Drm_Sprite unless we are going inactive and need to iterate
sprites. (and also fixes some formatting)

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix formatting
Chris Michael [Wed, 10 Dec 2014 17:14:25 +0000 (12:14 -0500)]
ecore-drm: Fix formatting

Summary: No functional changes, just formatting fixes

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix formatting
Chris Michael [Wed, 10 Dec 2014 17:13:33 +0000 (12:13 -0500)]
ecore-drm: Fix formatting

Summary: No functional changes, just formatting fixes

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add event to notify session activation state.
Seunghun Lee [Wed, 10 Dec 2014 17:12:21 +0000 (12:12 -0500)]
ecore-drm: Add event to notify session activation state.

Summary:
this event is to notify session activation state to compositor.
thus compositor can change composite state by this event.

Reviewers: devilhorns

Subscribers: torori, cedric

Differential Revision: https://phab.enlightenment.org/D1767

9 years agoecore-drm: Finish implementing touch events
Chris Michael [Wed, 10 Dec 2014 17:01:16 +0000 (12:01 -0500)]
ecore-drm: Finish implementing touch events

Summary: This fixes the touch events to actually send button down/up
when the proper touch event occurs

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Also update device calibration when output size changes
Chris Michael [Wed, 10 Dec 2014 17:00:22 +0000 (12:00 -0500)]
ecore-drm: Also update device calibration when output size changes

Summary: When the output size gets updated, it could end up changing
the device calibration (wrt input event transformation), so we should
update the device calibration also

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-evas-drm: Fix up error handling
Chris Michael [Wed, 10 Dec 2014 15:45:43 +0000 (10:45 -0500)]
ecore-evas-drm: Fix up error handling

Summary: Due to the change of order in creating outputs & inputs, we
need an additional goto error for destroying inputs if output_create
fails

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-evas-drm: Create inputs before we create outputs
Chris Michael [Wed, 10 Dec 2014 15:43:59 +0000 (10:43 -0500)]
ecore-evas-drm: Create inputs before we create outputs

Summary: Call function to create drm input devices Before we create
outputs. This is done so that when outputs Do get created, it will
update the input's known size of output for device coordinate
transformation

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Comment out debug line for unhandled touch_frame event
Chris Michael [Wed, 10 Dec 2014 15:43:28 +0000 (10:43 -0500)]
ecore-drm: Comment out debug line for unhandled touch_frame event

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Update pointer motion event coordinates
Chris Michael [Wed, 10 Dec 2014 15:19:54 +0000 (10:19 -0500)]
ecore-drm: Update pointer motion event coordinates

Summary: This fixes pointer motion events to be transformed based on
output coordinates

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Transform touch events into output coordinates
Chris Michael [Wed, 10 Dec 2014 15:17:54 +0000 (10:17 -0500)]
ecore-drm: Transform touch events into output coordinates

Summary: This updates touch events to be transformed into output
coordinates

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add internal function to set the output size in the input
Chris Michael [Wed, 10 Dec 2014 15:16:24 +0000 (10:16 -0500)]
ecore-drm: Add internal function to set the output size in the input
device structure

Summary: This is used to update all input devices as to the new output
size so that input device events can get their coordinates transformed
into output coords

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Update input devices with new output size
Chris Michael [Wed, 10 Dec 2014 15:15:57 +0000 (10:15 -0500)]
ecore-drm: Update input devices with new output size

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add output size to device structure and add internal
Chris Michael [Wed, 10 Dec 2014 15:15:10 +0000 (10:15 -0500)]
ecore-drm: Add output size to device structure and add internal
function for setting size

Summary: This is needed so that events for mouse, touch, etc can get
transformed into output coordinates

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix formatting
Chris Michael [Wed, 10 Dec 2014 15:04:22 +0000 (10:04 -0500)]
ecore-drm: Fix formatting

Summary: No functional changes, just formatting fixes

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix usage of EINA_LIST_FREE
Chris Michael [Wed, 10 Dec 2014 15:00:13 +0000 (10:00 -0500)]
ecore-drm: Fix usage of EINA_LIST_FREE

Summary: When we exit the backlight_init function, we should
stringshare_del the returned device strings from eeze properly

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix formatting
Chris Michael [Wed, 10 Dec 2014 14:59:48 +0000 (09:59 -0500)]
ecore-drm: Fix formatting

Summary: No functional changes, just fixing formatting

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Handle various touch events
Chris Michael [Wed, 10 Dec 2014 14:45:29 +0000 (09:45 -0500)]
ecore-drm: Handle various touch events

Summary: This adds code to handle various touch events (up, down,
motion) so that touchscreen devices can work with ecore_drm.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Remove libinput log print callback
Chris Michael [Wed, 10 Dec 2014 14:44:38 +0000 (09:44 -0500)]
ecore-drm: Remove libinput log print callback

Summary: This was causing crashes in libinput when trying to print out
via eina_log. Disable/Remove it. Let libinput print it's own logs.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoelua: always install cmake stuff for elua, even when not being built
Daniel Kolesa [Wed, 10 Dec 2014 15:25:00 +0000 (15:25 +0000)]
elua: always install cmake stuff for elua, even when not being built

9 years agoecore_drm: Setup correct linker flags for usage with systemd-login
Stefan Schmidt [Wed, 10 Dec 2014 14:09:44 +0000 (15:09 +0100)]
ecore_drm: Setup correct linker flags for usage with systemd-login

After the introduction of the non-systemd launcher the linking with systemd
enabled failed for me.

lib/ecore_drm/.libs/libecore_drm.so: undefined reference to `sd_pid_get_session'

Better we make sure that we actually setup the linker flags for the systemd use
case as well.

9 years agomodules/tgv: Switch from obsolete LZ4_uncompress function to LZ4_decompress_fast
Stefan Schmidt [Wed, 10 Dec 2014 13:47:42 +0000 (14:47 +0100)]
modules/tgv: Switch from obsolete LZ4_uncompress function to LZ4_decompress_fast

The function is declared as obsolete and in some newer headers it is no
longer defined. The new function is the same besides the name change.

9 years agoconfigure: Fix typo in systemd-login check
Stefan Schmidt [Wed, 10 Dec 2014 13:43:19 +0000 (14:43 +0100)]
configure: Fix typo in systemd-login check

9 years agoeet: Switch from obsolete LZ4_uncompress function to LZ4_decompress_fast
Stefan Schmidt [Wed, 10 Dec 2014 13:34:42 +0000 (14:34 +0100)]
eet: Switch from obsolete LZ4_uncompress function to LZ4_decompress_fast

The function is declared as obsolete and in some newer headers it is no
longer defined. The new function is the same besides the name change.

9 years agoecore-drm: fix passing wrong argument to open().
Seunghun Lee [Wed, 10 Dec 2014 13:04:32 +0000 (08:04 -0500)]
ecore-drm: fix passing wrong argument to open().

Summary: remove unintended argument.

@fix

Reviewers: devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1765

9 years agoecore-drm: restore VT to text mode when close tty.
Seunghun Lee [Wed, 10 Dec 2014 13:03:44 +0000 (08:03 -0500)]
ecore-drm: restore VT to text mode when close tty.

Summary: if ecore-drm failed to be run after changing tty mode, need to restore VT to text mode.

Test Plan:
- run enlightenment(drm backend) without systemd.
after changing tty graphics mode, if it's failed to be run by any reason,
you'll see black screen and coulnd't control session.

@fix

Reviewers: devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1764

9 years agoecore-drm: cleanup data when failed to setup tty.
Seunghun Lee [Wed, 10 Dec 2014 13:02:45 +0000 (08:02 -0500)]
ecore-drm: cleanup data when failed to setup tty.

Summary: added code to free string of tty's name when failed to setup tty.

@fix

Reviewers: devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1763

9 years agoecore_drm: Removed unused variables in ecore_drm_evdev
kabeer khan [Wed, 10 Dec 2014 13:01:19 +0000 (08:01 -0500)]
ecore_drm: Removed unused variables in ecore_drm_evdev

Summary: Signed-off-by: kabeer khan <kabeer.khan@samsung.com>

@fix

Reviewers: devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1762

9 years agoecore-drm: Added cleanup data code as a error handling of connect logind.
Seunghun Lee [Wed, 10 Dec 2014 13:00:17 +0000 (08:00 -0500)]
ecore-drm: Added cleanup data code as a error handling of connect logind.

Summary: need to clean also dbus, when failed open VT as a connection progress of logind.

@fix

Reviewers: devilhorns

Reviewed By: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1756

9 years agoecore-drm: Fix formatting.
Chris Michael [Wed, 10 Dec 2014 12:59:10 +0000 (07:59 -0500)]
ecore-drm: Fix formatting.

Summary: No functional changes, just formatting fixes

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoedge: fix memory leak and removed redundant return statement.
Srivardhan Hebbar [Wed, 10 Dec 2014 10:19:59 +0000 (11:19 +0100)]
edge: fix memory leak and removed redundant return statement.

Summary:
The name was freed after exit, so freeing it before and also freeing in a error case. There was a return statement after exit, thought it was redundant and removed.
@fix

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: devilhorns, cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1758

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecore_audio: fix memory leak in ecore_audio_pulse.
kabeer khan [Wed, 10 Dec 2014 10:19:18 +0000 (11:19 +0100)]
ecore_audio: fix memory leak in ecore_audio_pulse.

Summary: Signed-off-by: kabeer khan <kabeer.khan@samsung.com>

@fix

Reviewers: devilhorns, cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1761

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoedje: Edje_Edit - generate missed properties of 'text' block
Maksym Volodin [Wed, 10 Dec 2014 02:44:50 +0000 (03:44 +0100)]
edje: Edje_Edit - generate missed properties of 'text' block

Summary:
missed souce, text-source, repch, size_range, style are added.
TODO is added for filter property.

Reviewers: Hermet, raster, cedric

Subscribers: cedric, reutskiy.v.v

Differential Revision: https://phab.enlightenment.org/D1754

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoEvas GL: Fix direct rendering with client-side rotation
Jean-Philippe Andre [Tue, 9 Dec 2014 12:36:45 +0000 (21:36 +0900)]
Evas GL: Fix direct rendering with client-side rotation

There was a problem when checking whether the current surface
is compatible with direct rendering. In case of client-side
rotation (it's a flag set on the surface by the app), a surface
can be directly rendered even if the rotation is not 0.

But, before this patch, it was assumed that the surface was
current. Which doesn't make sense because make_current is
called by the pixel callback, from the application, and this
happens *after* we check for direct rendering.

As a consequence, it was not possible to mix directly rendered
surfaces with FBO-based ones, and use client-side rotation.

This patch should solve that issue.

9 years agoevil: add getpwnam() function
Vincent Torri [Tue, 9 Dec 2014 19:35:15 +0000 (20:35 +0100)]
evil: add getpwnam() function

@feature

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoecoreodrm - fix build if systemd is not enabled
Carsten Haitzler (Rasterman) [Tue, 9 Dec 2014 23:54:02 +0000 (08:54 +0900)]
ecoreodrm - fix build if systemd is not enabled

9 years agoecore-drm: Merge port to libinput
Chris Michael [Tue, 9 Dec 2014 20:16:27 +0000 (15:16 -0500)]
ecore-drm: Merge port to libinput

Merge branch 'devs/devilhorns/libinput'

Summary: These commits port the current ecore_drm input code to use
libinput for handling of devices, input events, etc

@feature

9 years agoecore-drm: Port ecore_drm to use libinput
Chris Michael [Tue, 9 Dec 2014 20:13:49 +0000 (15:13 -0500)]
ecore-drm: Port ecore_drm to use libinput

Summary: This ports the input code of ecore_drm to make use of
libinput for handling of devices, events, etc

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix ecore_drm_launcher_device_close to not cause segfaults
Chris Michael [Tue, 9 Dec 2014 20:12:36 +0000 (15:12 -0500)]
ecore-drm: Fix ecore_drm_launcher_device_close to not cause segfaults

Summary: We can only call _ecore_drm_logind_device_close if we have an
actual device string. It is possible that this Could get called with
NULL, and thus cause a crash.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Cleanup structures in private header
Chris Michael [Tue, 9 Dec 2014 20:11:46 +0000 (15:11 -0500)]
ecore-drm: Cleanup structures in private header

Summary: With the port to libinput, some fields in the input & evdev
structures are not needed anymore, so this commit cleans those up.

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add API function for eeze_udev_find_by_subsystem_sysname
Chris Michael [Tue, 9 Dec 2014 20:10:43 +0000 (15:10 -0500)]
eeze: Add API function for eeze_udev_find_by_subsystem_sysname

Summary: This adds an API function to find a device by
subsystem_sysname. This is needed for the port of ecore_drm to use
libinput.

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoeeze: Add API function declaration for find_by_subsystem_sysname
Chris Michael [Tue, 9 Dec 2014 20:09:25 +0000 (15:09 -0500)]
eeze: Add API function declaration for find_by_subsystem_sysname

Summary: This adds an API function declaration for
eeze_udev_find_by_subsystem_sysname function which is needed for the
porting of ecore_drm to use libinput

@feature

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Include libinput header in drm private header
Chris Michael [Tue, 9 Dec 2014 15:00:29 +0000 (10:00 -0500)]
ecore-drm: Include libinput header in drm private header

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Add libinput dependency check
Chris Michael [Tue, 9 Dec 2014 14:58:27 +0000 (09:58 -0500)]
ecore-drm: Add libinput dependency check

Summary: ecore-drm will now require libinput for handling input
devices, so this commit adds a configure check for libinput

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix formatting
Chris Michael [Tue, 9 Dec 2014 14:52:24 +0000 (09:52 -0500)]
ecore-drm: Fix formatting

Summary: No functional changes, just fixing formatting

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: Fix error message typo
Chris Michael [Tue, 9 Dec 2014 14:50:59 +0000 (09:50 -0500)]
ecore-drm: Fix error message typo

Summary: This just fixes a typo in the error message string

@fix

Signed-off-by: Chris Michael <cp.michael@samsung.com>
9 years agoecore-drm: added drm launcher that is allow to determine whether to launch with login...
Seunghun Lee [Tue, 9 Dec 2014 14:36:42 +0000 (09:36 -0500)]
ecore-drm: added drm launcher that is allow to determine whether to launch with logind or root privilege.

Summary:
- allow to launch drm backend without systemd-logind with root privilege.
- allow to open drm device node via logind, not directly open it, in case exist systemd-logind.
- fixes issue that couldn't switch session, because ecore-drm couldn't drop master to drm device with no permission. (allow to switch session appropriate.)

Reviewers: gwanglim, devilhorns

Subscribers: torori, cedric

Differential Revision: https://phab.enlightenment.org/D1704

9 years agoethumb: Fixing eina_init to eina_shutdown.
Srivardhan Hebbar [Tue, 9 Dec 2014 14:24:38 +0000 (09:24 -0500)]
ethumb: Fixing eina_init to eina_shutdown.

Summary:
In the end we should call eina_shutdown, but by mistake eina_init is being called. So have corrected that.

@fix

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1751

9 years agoevas_xcb_outbuf: fix memory leak
kabeer khan [Tue, 9 Dec 2014 14:22:41 +0000 (09:22 -0500)]
evas_xcb_outbuf: fix memory leak

Summary: This fixes a leak in the xcb engine where the Outbuf_Region
was not being freed.

@fix

Signed-off-by: kabeer khan <kabeer.khan@samsung.com>
Reviewers: devilhorns

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1750

9 years agoeina: removed multiple eina_init call.
Srivardhan Hebbar [Tue, 9 Dec 2014 10:16:42 +0000 (11:16 +0100)]
eina: removed multiple eina_init call.

Summary:
I observed that eina_init() was called twice in this function and eina_shutdown() once. So thought this might be a mistake, and so have cleaned it.

Signed-off-by: Srivardhan Hebbar <sri.hebbar@samsung.com>
Reviewers: devilhorns, cedric

Reviewed By: cedric

Subscribers: cedric

Differential Revision: https://phab.enlightenment.org/D1748

Signed-off-by: Cedric BAIL <cedric@osg.samsung.com>
9 years agoEvas GL: Fix scissors with direct rendering
Jean-Philippe Andre [Tue, 9 Dec 2014 09:35:35 +0000 (18:35 +0900)]
Evas GL: Fix scissors with direct rendering

If an app calls glDisable(SCISSORS) and uses direct rendering,
then the DR scissors were dropped and so glClear would erase
the contents of the entire canvas, instead of being restricted
to the image object.

9 years agoEvas GL: Fix current FBO after make_current(0, 0)
Jean-Philippe Andre [Tue, 9 Dec 2014 09:02:20 +0000 (18:02 +0900)]
Evas GL: Fix current FBO after make_current(0, 0)

Example scenario:
- Create a direct rendered Evas GL 'sfc' 'ctx'
- Create a PBuffer dummy surface, make it current
- Do some stuff
- Make current (NULL, NULL) to go back to no target
- Make current (sfc, ctx)
--> glClear() will not render anything on screen

Reason:
The current FBO binding is still set to the implicit FBO
bound to the PBuffer surface (it could be any surface, really).

9 years agoedje/edje_cc: don't limit the bezier cubic minium value to 0.
ChunEon Park [Tue, 9 Dec 2014 07:59:58 +0000 (16:59 +0900)]
edje/edje_cc: don't limit the bezier cubic minium value to 0.

No need to limit the minimum value to 0.
let user have a choice.

9 years agoecore/animator: improve the doc description.
ChunEon Park [Tue, 9 Dec 2014 07:51:08 +0000 (16:51 +0900)]
ecore/animator: improve the doc description.

9 years agoecore/animator: code refactoring.
ChunEon Park [Tue, 9 Dec 2014 07:43:23 +0000 (16:43 +0900)]
ecore/animator: code refactoring.

renamed to more meaningful name.
don't stupid computaion twice.